1

The erosion rate and previous extent of interior layered deposits on Mars revealed by

obstructed landslides

Supplementary Information

P. M. Grindrod and N. H. Warner

Figure DR1. Context map of the study area in (a) Valles Marineris and (b) Ophir Chasma. Black

boxes show the location of figures in the main text and Supplementary Material. Large-scale

interior layered deposits (ILDs) are highlighted throughout the wider Valles Marineris region.

GSA DATA REPOSITORY 2014287

2

Figure DR2. Typical Valles Marineris landslides (a) with and (b) without obstructions. (a) CTX

image of the landslide Ophir Labes and topographic profile showing landslide obstruction by a

bedrock peak that has not been removed. (b) CTX image of the landslide Coprates Labes and

topographic profile showing landslide frontal region with no obstruction.

3

Figure DR3. Perspective view of an obstructed landslide. The frontal scarp is 1 – 2 km from and

mimics the shape of the ILD. Perspective view made from a CTX stereo DTM and image. Also

shown (inset) is a close-up view of the frontal scarp.

Figure DR4. Schematic diagram, not to scale, showing the proposed formation scenario. (a)

Landslide deposits flow away from their source region and are obstructed by the ILDs. Shown are

both minimum and maximum representative sizes of ILDs that could have caused such an

obstruction. (b) After landslide formation the ILDs have reduced in size, causing limited slumping

at the toe of the landslide, but with landslide material volumes being conserved.

4

Figure DR5. Crater count results for the ILDs. Plot is binned cumulative crater frequency

histogram for all craters greater than 50 m diameter. PF is the production function (Ivanov, 2001)

and CF is the chronology function (Hartmann and Neukum, 2001).

Interior Layered Deposits Volume Loss

We estimate the minimum and maximum volume of material lost from the ILDs by

assuming that the retreat at the front of the landslides is indicative of volume loss elsewhere. To

estimate the minimum volume of ILD material that must have been removed, we calculated the area

defined by the current landslide frontal scarp and multiplied by either the path length or

circumference of the ILD boundaries. Some of the landslide toe (the highest point) has likely fallen

into the void left by the ILD removal, but the overall effect on the results will be small as volume

will be conserved. Therefore if material has been lost just from the northern slopes of the two ILDs,

then the minimum volume removed is ~0.4 x 103 km

3; if however material has been lost from the

entire circumference of the two ILDs then the minimum volume removed is ~1.2 x 103 km

3. If

material has been removed from the entire ILD, rather than just at the base close to the landslide,

then we can calculate maximum estimates of material loss. In this case we cropped a 50 m/px

5

HRSC DTM mosaic to the ILD outline, then added 500 m of elevation and calculated the volume

difference. Therefore if material has been lost just from the northern boundary of both ILDs, then

the maximum volume removed is ~1.5 x 103 km

3; if however material has been lost from the entire

circumference of the two ILDs then the maximum volume removed is ~5.8 x 103 km

3.

Table DR1. Results of landslide crater counts.

Landslide No. Area

(km2)

N

(craters in age)

Fit Range

(km) N(1) N(10)

Model Age

(Ma)

1 157 130 0.1 – 0.48 2.1 x 10-4

3.0 x10-6

423 +55/-58

2 360 276 0.1 – 0.45 1.8 x10-4

2.6 x10-6

374 +34/-37

3 308 228 0.1 – 0.45 1.4 x10-4

2.8 x10-6

397 +40/-42

4 312 173 0.1 – 0.40 1.5 x10-4

2.1 x10-6

308 +35/-38

5 263 115 0.1 – 0.35 1.1 x10-4

1.6 x10-6

225 +31/-34

6 509 197 0.1 – 0.48 1.1 x10-4

1.5 x10-6

216 +23/-24